Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

“Bloody hell,” muttered Finn, his Irish brogue thick with fury as he slammed the newspaper onto the bar. *“A pale imitation of traditional fare.”* His own mother’s recipes, denigrated. He felt a throbbing pulse in his temples.

He poured himself a hefty whiskey, the amber liquid sloshing over the rim of the glass. The usually inviting aroma of the pub, of warm beer and brewing stew, now seemed like a taunt. The room was empty.

He saw his reflection in the mirror: flushed face, eyes narrowed. It was all a lie, every single word. The critic didn't know anything about good food.

He threw the glass against the wall, smashing it to a thousand pieces.

Emotion: resentful

Cluster: Shame / Guilt
PC1 (Valence): -1.47 Negative
PC2 (Disposition): 1.50

Role in Research

This story is one of 1,000 stories generated for the emotion resentful. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 resentful stories, after denoising with neutral dialogue baselines, produces the resentful emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the resentful vector is projected through the unembedding matrix.

Promoted:
S0.485
aggravated0.263
愤0.261
👎0.245
L0.241
Suppressed:
own-0.318
soon-0.225
随后-0.225
serendip-0.219
🤩-0.210